Document généré le 6 oct. 2021 02:47

Géographie physique et Quaternaire

The Paleoecological Record of 6 ka BP Climate in the Canadian Prairie Provinces Données paléoécologiques sur le climat des provinces des Prairies à 6 ka BP Palaoökologische Belege über das Klima in den kanadischen Prärie-Provinzen um 6 ka v.u.Z. Robert E. Vance, Alwynne B. Beaudoin et Brian H. Luckman

La paléogéographie et la paléoécologie d’il y a 6000 ans BP au Canada Résumé de l'article Paleogeography and Paleoecology of 6000 yr BP in Canada La synthèse des études paléoécologiques montrent qu'à 6 ka le maximum Volume 49, numéro 1, 1995 d'aridité, qui caractérisait l'Holocène inférieur, avait été atteint, mais le climat demeurait plus chaud et plus sec que maintenant dans toute la région. Les URI : https://id.erudit.org/iderudit/033031ar différentes limites des arbres étaient plus élevées qu'aujourd'hui et la DOI : https://doi.org/10.7202/033031ar superficie des glaciers alpins étaient plus petite, les niveaux lacustres étaient moins élevés dans la plus grande partie des plaines intérieures et les écozones des Prairies et de la forêt boréale s'étendaient plus au nord qu'aujourd'hui. Les Aller au sommaire du numéro feux de forêt étaient plus fréquents à 6 ka qu'actuellement, ce qui a favorisé la migration du pin gris (Pinus banksiana) vers l'ouest à travers la forêt boréale et accru la superficie de la prairie dans les zones de forêts boréale et alpine. Les Éditeur(s) essais faits dans le but de mesurer l'ampleur des différences de températures et de précipitations à 6 ka ont donné des résultats variables, mais montrent que Les Presses de l'Université de Montréal la température était de 0,5 à 1,5°C plus élevé que maintenant (avec une température estivale jusqu'à 3°C plus élevée) et les précipitations annuelles ISSN moyennes étaient de 65 mm inférieures (avec des précipitations estivales de 50 0705-7199 (imprimé) mm inférieures) à maintenant. La nature et le degré des changements laissent 1492-143X (numérique) croire qu'une forte circulation atmosphérique zonale, semblable à celle des années 1930, mais plus nordique, existait à 6 ka. Découvrir la revue

Citer cet article Vance, R. E., Beaudoin, A. B. & Luckman, B. H. (1995). The Paleoecological Record of 6 ka BP Climate in the Canadian Prairie Provinces. Géographie physique et Quaternaire, 49(1), 81–98. https://doi.org/10.7202/033031ar

Tous droits réservés © Les Presses de l'Université de Montréal, 1995 Ce document est protégé par la loi sur le droit d’auteur. L’utilisation des services d’Érudit (y compris la reproduction) est assujettie à sa politique d’utilisation que vous pouvez consulter en ligne. https://apropos.erudit.org/fr/usagers/politique-dutilisation/

Cet article est diffusé et préservé par Érudit. Érudit est un consortium interuniversitaire sans but lucratif composé de l’Université de Montréal, l’Université Laval et l’Université du Québec à Montréal. Il a pour mission la promotion et la valorisation de la recherche. https://www.erudit.org/fr/ Géographie physique et Quaternaire, 1995, vol. 49, n° 1, p. 81-98, 8 fig., 1 tabl..

THE PALEOECOLOGICAL RECORD OF 6 KA BP CLIMATE IN THE CANADIAN PRAIRIE PROVINCES*

Robert E. VANCE, Alwynne B. BEAUDOIN and Brian H. LUCKMAN, respectively: Geological Survey of Canada, 3303- 3310St. NW, Calgary, Alberta T2L 2A7; Archaeological Survey, Provincial Museum of Alberta, 12845-102nd Ave., Edmon­ ton, Alberta T5N 0M6; Department of Geography, University of Western Ontario, London, Ontario N6A 5C2.

ABSTRACT Synthesis of available RÉSUMÉ Données paléoécologiques sur ZUSAMMENFASSUNG Palaoôkologische paleoecological studies in the prairie prov­ le climat des provinces des prairies à 6 ka. Belege ùber das Klima in den kanadischen inces of Canada indicates that although the BP La synthèse des études paléoécologi­ Prârie-Provinzen um 6 ka v.u.Z. Die peak in postglacial aridity that character­ ques montrent qu'à 6 ka le maximum d'ari­ Synthèse der verfùgbaren palâoôkolo- ized early Holocene climate of the western dité, qui caractérisait l'Holocène inférieur, gischen Studien in den Prârie-Provinzen foothills and plains had passed, conditions avait été atteint, mais le climat demeurait Kanadas zeigt, daf3 das Klima in der ganzen remained warmer and drier than present plus chaud et plus sec que maintenant dans Region um etwa 6000 Jahre v.u.Z. warmer throughout the region ca. 6000 yr BR Com­ toute la région. Les différentes limites des und trockener als gegenwârtig blieb, wenn pared to today, treeline elevations were arbres étaient plus élevées qu'aujourd'hui auch das Maximum an postglazialer higher and alpine glaciers were reduced in et la superficie des glaciers alpins étaient Trockenheit, welches im frùhen Holozàn das size in the Rocky Mountains, lake levels plus petite, les niveaux lacustres étaient Klima der westlichen Gebirgsauslâufer und were lower over much of the Interior Plains, moins élevés dans la plus grande partie Ebenen charakterisierte, vorùber war. and the grassland and boreal forest des plaines intérieures et les écozones des Verglichen mit heute waren die Baum- ecozones extended north of their present prairies et de la forêt boréale s'étendaient grenzen hôher und die alpinen Gletscher in positions. Forest fires were more prevalent plus au nord qu'aujourd'hui. Les feux de den Rocky Mountains kleiner, die Seehôhen ca. 6000 yr BP than they are today, aiding forêt étaient plus fréquents à 6 ka qu'ac­ waren niedriger im grôBten Teil der inneren westward migration of jack pine (Pinus tuellement, ce qui a favorisé la migration du Ebenen, und die Ôkozonen von Grasland banksiana) through the boreal forest and pin gris (Pinus banksiana) vers l'ouest à und nôrdlichem WaId erstreckten sich weiter increasing the area occupied by grassland travers la forêt boréale et accru la superfi­ nôrdlich als heute. Waldbrânde traten um in boreal and montane forest regions. At­ cie de la prairie dans les zones de forêts etwa 6000 v.u.Z. hâufiger als gegenwârtig tempts to quantify the magnitude of 6 ka boréale et alpine. Les essais faits dans le auf und haben so die Westwàrtswanderung temperature and precipitation differences but de mesurer l'ampleur des différences von Graukiefer (Pinus banksiana) durch den have produced variable results, but sug­ de températures et de précipitations à 6 ka nôrdlichen WaId begùnstigt und die gest that mean annual temperature was ont donné des résultats variables, mais Graslandflàchen in den Gebieten des 0.50C to 1.50C higher than today (summer montrent que la température était de 0,5 à nôrdlichen und alpinen Waldes vergrôBert. temperature may have been up to 3°C 1,5°C plus élevé que maintenant (avec une Versuche, den Umfang der Unterschiede in higher) and mean annual precipitation was température estivale jusqu'à 3°C plus éle­ Temperatur und Niederschlàgen um 6 ka reduced by 65 mm (or summer precipita­ vée) et les précipitations annuelles moyen­ zu bestimmen, ergaben wechselnde tion was reduced by 50 mm), compared to nes étaient de 65 mm inférieures (avec des Ergebnissse, doch zeigen sie, daB die present. The nature and scale of these précipitations estivales de 50 mm inférieu­ durchschnittliche jâhrliche Temperatur 0.5° changes suggests that a vigorous zonal res) à maintenant. La nature et le degré C bis 1.50C hôher als gegenwârtig war (die atmospheric circulation pattern, similar to des changements laissent croire qu'une Sommertemperatur dûrfte bis zu 3°C hôher that of the 1930s but shifted northward, pre­ forte circulation atmosphérique zonale, sem­ gewesen sein) und die durchschnittlichen vailed at 6 ka. blable à celle des années 1930, mais plus jâhrlichen Niederschlâge waren um 65 mm nordique, existait à 6 ka. geringer (oder die Sommer-Niederschlàge waren um 50 mm geringer) verglichen mit heute. Die Natur und der Grad dieser Wechsel lassen vermuten, daB eine krâftige zonale atmosphàrische Strômung, àhnlich der von den 1930 iger Jahren aber nôrdlicher gelagert, um 6 ka vorherrschte.

Manuscrit reçu le 10 juin 1994; manuscrit révisé accepté le 7 novembre 1994 * Geological Survey of Canada Contribution No. 52193 82 R.E. VANCE, A.B. BEAUDOIN and B.H. LUCKMAN

INTRODUCTION less landscape of gently undulating grassland. In fact, the three prairie provinces embody a great deal of geomorphic, Shifts in the position of regional vegetation belts and climatic, and ecological variability. Large scale physiographic changes to regional surface hydrology are among the clear­ features, shaped by Cretaceous and Tertiary bedrock, form est indicators of past climate change. Over the last twenty- a series of prairie 'steps' across the southern half of the five years, an extensive network of paleoecological study three provinces (Klassen, 1989). The easternmost step, the sites focused on these topics has been developed in Mani­ Manitoba Escarpment, separates the Manitoba Plain (ca. toba, Saskatchewan, and Alberta; a network of sufficient 250 m asl) from the Saskatchewan Plain (ca. 400-800 m scope to allow a preliminary assessment of paleoclimatic asl) which, in turn, is separated from the Alberta Plain (ca. conditions across the Canadian Prairies. A number of 800-1000 m asl) by the Missouri Coteau. The Alberta Plain paleoclimatic indices may be consulted to reconstruct cli­ grades into the eastern slope foothills of the Rocky Moun­ matic conditions ca. 6000 yr BP, including several pollen tains which rise steeply to the continental divide at ca. stratigraphie studies from all ecological zones, investigation 3000 m asl) In the northwest, the Alberta Plateau (ca. 600- of high altitude glacier and treeline dynamics in the Rocky 900 m asl) is dissected by the Peace River Lowland (ca. Mountains, as well as a number of diverse paleolimnological 300-600 m asl). The northeastern portion of the prairie studies. The purpose of this synthesis is to describe, both provinces is part of the Precambrian Shield, whose surface qualitatively and quantitatively, how climatic conditions 6000 expression consists of extensive poorly drained, low relief years ago differed from those of today. This period is of plains and plateaus dotted with numerous lakes and interest now because the 6 ka time slice has been chosen wetlands. This topography contrasts markedly with the south­ by the NATO sponsored Paleoclimate Model lntercomparison ern prairies where low relief is interrupted by several promi­ Project as one point for testing Global Climate Model output nent uplands (e.g., Turtle Mountain, Wood Mountain, and with the paleoecological record (Jette, 1995). the Cypress Hills). Proxy indicators provide indirect evidence of past cli­ The prairie provinces are drained to the north by major mate, but the signal may be muted or filtered by interaction tributaries of the Mackenzie system flowing to the Arctic with other environmental factors. As a result, it is rarely Ocean, the Churchill and Saskatchewan-Nelson River sys­ possible to use a single proxy data record directly for cli­ tems draining east to Hudson Bay, and to the south by matic reconstruction. Moreover, a specific proxy indicator Mississippi tributaries flowing to the Gulf of Mexico. Many may reflect different climatic parameters in differing areas of these major river systems have their headwaters in the or ecozones. To date, most research has focused on biotic western Cordillera. Several rivers (e.g., the Bow and Sas­ proxy indicators, producing a wealth of paleoecological in­ katchewan) are glacier-fed and have cut deeply incised formation from remains preserved in lake sediment, includ­ valleys across the Interior Plains. Thus the hydrological ing indicators of terrestrial conditions (e.g., pollen) and system integrates inputs from the Rocky Mountains and limnological indicators (e.g., diatoms). Cases where several Interior Plains and responds to precipitation and glacier indicators have been studied at a single site are particularly mass changes in the headwaters as well as changes in valuable because they provide an opportunity to link the precipitation and evaporation on the prairie surface. environmental history of the terrestrial and limnological sys­ The size and physiographic diversity of the prairie prov­ tems. inces makes it impossible to specify climatic parameters for Ritchie (1976, 1985, 1987, 1989) has summarized much the entire region. However, broad scale features of western of the existing pollen data in western Canada as part of a interior climate are dictated mainly by the interaction of broad regional history of the major vegetation zones of Arctic and Pacific air masses (although summers in the Canada. Anderson et al. (1989) discussed paleoecological southeast are influenced by warm, moist air masses origi­ data (mainly pollen) pertinent to the Hypsithermal interval in nating in the Gulf of Mexico). During winter months, when western Canada (ca. 9000-4000 yr BP), and Ritchie and the core of westerly atmospheric circulation typically mi­ Harrison (1993) outlined Holocene vegetation and lake level grates south of the Canadian border, incursions of Pacific dynamics in western Canada, again relying mainly on pol­ air are infrequent and the area is often blanketed by cold, len data. Although providing valuable overviews, these sur­ dry Arctic air. With the northward movement of the westerly veys are neither comprehensive nor focus specifically on core during spring and summer, Pacific air masses occupy the 6 ka interval. The main source of proxy data cited in this the area more frequently, particularly in the south. This shift paper is also pollen, but these data are supplemented by in air mass dominance produces pronounced seasonal ex­ information on glacier dynamics, macrofossil evidence of tremes in temperature and precipitation. For the period 1930-1980, mean January temperatures in southern re­ high altitude treeline dynamics, sedimentologic and biotic 0 evidence of lake-level changes, and pédologie data related gions have been about -10 C and mean July temperatures about +200C, whereas in the north seasonal extremes vary to past aeolian and alluvial activity, in an attempt to bring 0 together as many published and unpublished sources of between -25°C and +15 C (Longley, 1972; Atmospheric Environment Service, 1982). Mean annual precipitation for data pertinent to the 6 ka interval as possible. the same period is variable, with peak values registered in SETTING west central Alberta (600 mm) and south central Manitoba (500 mm), and a low of 300 mm recorded in southeastern The term 'prairie provinces' (Fig. 1) invokes an Alberta and southwestern Saskatchewan. Measurements inapproriate image of vast tracts of an essentially feature­

Géographie physique et Quaternaire, 49(1), 1995 THE PRAIRIE PROVINCES 83

Mountains

ABB'93 200 km

FIGURE 1. Generalized ecological zones (based on Rowe, 1972) Zones écologiques généralisées (fondée sur Rowe, 1972) et and major drainage systems of the Canadian prairie provinces. principaux réseaux de drainage des provinces des prairies. Arbres Major tree species characteristic of each ecological zone: transi­ dominants caractéristiques de chacune des zones : transition entre tional coniferous forest and forest tundra — white spruce (Picea la forêt de conifères et la toundra forestière — épinette blanche glauca), black spruce (Picea mariana) and tamarack (Larix laricina); (Picea glauca), épinette noire (Picea mariana) et mélèze (Larix boreal forest — white spruce, black spruce, balsam fir {Abies laricina); forêt boréale — épinette blanche, épinette noire, sapin balsamea), jack pine (Pinus banksiana, birch (Betula papyrifera) baumier (Abies balsamea), pin gris (Pinus banksiana), bouleau and aspen (Populus tremuloides); parkland — open woodland with (Betula papyrifera) et peuplier faux-tremble (Populus tremuloides); trembling aspen; mixed forest — eastern white pine (Pinus strobus), tremblaie —parc; forêt mixte —pin blanc (Pinus strobus) bouleau, birch, aspen, Manitoba maple (Acer negundo), and oaks (Ouercus peuplier faux-tremble, érable négundo (Acer negundo) et chêne spp.). In the mountains, main tree species are lodgepole pine (Quercus spp.). En montagne, les principales espèces sont le pin (Pinus contorta var. Iatifolia) at lower elevations, Engelmann spruce tordu (Pinus contorta var Iatifolia) en basse altitude, !'épinette des (Picea engelmannii) and subalpine fir (Abies lasiocarpa) at higher montagnes Engelmann (Picea engelmannii) et le sapin subalpin elevations. Stars indicate locations of key paleoecological study (Abies lasiocarpa), en haute altitude. sites.

are not available for high elevation sites in the mountains, general course of upper westerly atmospheric flow. The two but Janz and Storr (1977) estimate precipitation exceeds most extensive vegetation types are the semi-arid grass­ 750 mm/yr for most areas above treeline. About 70% of the land in the southern area and the boreal forest in the north annual precipitation in the Interior Plains falls between April (Rowe, 1972; Fig. 1). These two ecozones are separated and September (Laycock,1972). Daily variability is also an by the transitional aspen parkland, a mosaic of grassland important component of western interior climate, with chinook and aspen groves whose northern limit is the vegetational winds (heralding incursions of Pacific air) causing rapid expression of the mean position of the winter Arctic front temperature increases in southwestern regions in winter (Bryson, 1966). The forest-tundra ecotone bounds the boreal and conversely, southward movement of Arctic air bringing forest to the northeast, occupying the Hudson Bay lowlands cold temperatures to the prairie provinces in all seasons, of northern Saskatchewan and Manitoba. In the west, el­ particularly in winter months, when temperatures of -400C evation is the primary influence on vegetation, with montane are not uncommon. coniferous forests in the foothills and lower elevations of The prairie provinces span several major ecological zones the Rocky Mountains, and alpine tundra upslope in the which, outside the mountains, trend NW-SE following the mountains.

Géographie physique el Quaternaire. 49(1), 1995 84 R.E. VANCE, A.B. BEAUDOIN and B.H. LUCKMAN

CHRONOLOGY 3) some important elements of the regional vegetation are palynologically almost "silent", either because of poor pol­ Precise identification of the 6 ka interval is not len preservation {e.g., Populus, Mott, 1978), inherent low staightforward for several reasons. Radiocarbon dating, the pollen productivity, or entomophily (Moore et ai, 1991). most frequently used technique for chronological control, Moreover, because of long distance transport and differen­ has an analytical uncertainty associated with each age tial deposition and preservation, pollen diagrams integrate determination, typically between ±75 to ±200 years. Addi­ data from an unknown area surrounding the site. Not sur­ tional errors are introduced depending upon the location prisingly, sites located near present-day ecotones show the and type of material dated. These are particularly signifi­ most clear-cut evidence of past climatic change because cant when bulk sediment samples from areas underlain they are situated at present day transitions between eco­ with carbonaceous shale, lignite, and limestone, for exam­ logical and climatic zones. Therefore the "broad-brush" pic­ ple the eastern slopes of the Rocky Mountains (Hickman ture of environmental conditions available through pollen and Schweger, 1993; White et al., 1985) and the southern analysis, although informative in monitoring movement of prairie grassland (Barnosky et al., 1987; Mott, 1973), are ecological zone boundaries and major vegetational changes, dated by conventional means. Accelerator mass needs, where possible, to be supplemented with data from spectrometry (AMS) radiocarbon dating of selected materi­ other proxy indicators to produce detailed paleoclimatic re­ als, primarily terrestrial plant macrofossils, is the most ef­ constructions. fective way of circumventing this problem (MacDonald et al., 1987,1991 ), but has not yet been used in many studies. A) NORTHERN FOREST TUNDRA Regardless of the dating technique, the 6000 yr BP interval The northern forest tundra on the northeastern margin of itself is seldom specifically dated. Hence, determination of the prairie provinces (Fig. 1) was deglaciated later than the the 6 ka time slice typically involves interpolation from other rest of the western interior, then inundated by Glacial Lake (often few) dated horizons, usually when deposition rates Agassiz and subsequently the Tyrrell Sea (Dredge and are poorly understood. Mazama ash, deposited 6845±50 yr Cowan, 1989a, b). Therefore much of the land was not BP (Bacon, 1983), is a readily recognizable temporal marker available for vegetation colonization until 6 ka. There are in the southwestern portion of the prairie provinces (Sarna- only two pollen records (Ennadai Lake and Lynn Lake) from Wojcicki et al., 1983) and in the mountains, but of course this region (Fig. 2), and both begin about 6000 yr BP On does not clearly identify the 6 ka horizon. Moreover, the the basis of relic podzols, developed in areas that are existence of possible earlier Mazama eruptions could cause presently tundra, and pollen analysis, Nichols (1967) in­ confusion (White and Osborn, 1992; Zoltai, 1989). Besides ferred that the northern margin of the boreal forest ex­ these analytical and sample errors, there is a significant tended further northward at 6 ka than it does today, sug­ discrepancy between radiocarbon years and calendar years gesting warmer summer temperatures than present. in the middle Holocene. Due to variations in the production of 14C in the upper atmosphere, a radiocarbon date of 6000 B) BOREAL FOREST yr BP is actually equivalent to 6850 calendar years (Stuiver Pollen records from within the boreal forest are domi­ and Becker, 1993). Because radiocarbon dates are typically nated by the main boreal taxa {Pinus, Picea, Alnus, Betula) not corrected in the literature and there is considerable and subtle changes in the structure or composition of the potential for errors related to the laboratory procedures and forest are difficult to discern. Pollen records from sites within sample type, the 6 ka time slice in this synthesis is broadly the central boreal forest are relatively complacent during defined as the interval between 5000 and 7000 radiocarbon the 6 ka interval (MacDonald, 1984, 1987; Mott, 1973; yr BP. Nichols, 1969; Kuhry et ai, 1992; Matthews, 1980; Ritchie, 1976; Ritchie and Hadden, 1975; Vance, 1986a; Wilson, THE PALEOECOLOGICAL RECORD 1984). Pollen frequency changes in these diagrams relate to migration of species {e.g., jack pine and alder) and forest The regional pollen stratigraphie record is emphasized in fires. The most significant change in middle Holocene boreal this synthesis because more pollen records are available forest pollen assemblages is an increase in pine. In western than any other type of proxy data in the prairie provinces Manitoba, Pinus increases from <5% to >40% abruptly at (Fig. 2 and 3). Although 6 ka coverage is uneven, with more about 5900 yr BP (Ritchie, 1976). In central Saskatchewan sites in Alberta than Saskatchewan and Manitoba com­ (Cycloid Lake) the increase occurs at 6000 yr BP (Mott, bined and the majority situated in boreal forest and moun­ 1973). Further west, the increase in pine pollen is more tain environs, the data set is adequate for a preliminary gradual {e.g., Lofty Lake, Lichti-Federovich, 1970, Fig. 4), assessment of past vegetation dynamics. However, pollen and pine is not locally abundant until about 4500 yr BP. This analysis is a relatively coarse tool for climate reconstruction westward migration of jack pine was probably encouraged because: 1) many taxa are identifiable only to the family or by frequent fires caused by middle Holocene warmth (Wilson, genus level (Faegri et al., 1989), resulting in loss of indi­ 1984). vidual species' autecological information, 2) pollen records in western Canada are typically dominated by a few prolific Generally, boreal forest/foothills sites in the northwest­ anemophilous taxa, primarily conifers (MacDonald and ern extreme of the prairie provinces do not display substan­ Ritchie, 1986), that are capable of "drowning out" minor tial changes in pollen frequencies during the 6 ka interval constituents and masking subtle vegetation changes, and (White and Mathewes, 1982, 1986), although subtle changes

Géographie physique et Quaternaire. 49(1), 1995 THE PRAIRIE PROVINCES 85

in indicator taxa suggest the onset of a wetter and cooler River district, now a perennial lake, was a seasonal pond at climatic regime, compared to that of the early Holocene 6 ka. Since the basin was dry in the early Holocene, 6 ka (Anderson et al., 1989; Schweger and Hickman, 1989). conditions were cooler and moister than previously, but White and Mathewes (1982), on the basis of relatively high warmer and possibly drier than present. At nearby Boone Cheno-Am, Typha latifolia, Artemisia, and Gramineae pol­ Lake, increases in pine pollen around 6 ka suggest that the len frequencies, suggested that Fiddler's Pond in the Peace forests were more fire-prone than they are now (White and

FIGURE 2. Paleoenvironmental records with chronologic controlI Données paléoenvironnementales sur les provinces des prairies et from the prairie provinces and adjacent areas that span the 6 kai les régions avoisinantes avec chronologie bien établie de la période interval (7000-5000 yr BP). Most records concentrate on polleni de 6 ka (de 7000 à 5000 BP). La plus grande partie des données data, some include or focus on macrofossils, geochemistry, orr concernent la palynologie et certaines comprennent des données diatoms. For more detail on the Alberta coverage, see Beaudoini sur les macrofossiles, la géochimie et les diatomées. (1993). 1. Ennadai Lake (Nichols, 1967), 2. John Klondike Bog (Matthews, (Luckman and Kearney, 1986), 34. Tonquin Pass (Kearney and 1980), 3. Lac Ciel Blanc (MacDonald, 1984), 4. Wild Spear Lake3 Luckman, 1983), 35. Lorraine Lake (Bear, 1989), 36. Maligne Lake (MacDonald, 1987), 5. Eaglenest Lake (Vance, 1986a), 6. Snow- kettle (Kearney and Luckman, 1987), 37. Porcupine Mountain shoe Lake (MacDonald, 1987), 7. Lynn Lake (Nichols, 1967), 8. (Nichols, 1969), 38. Buffalo Lake (Schweger and Hickman, 1989), Yesterday Lake (MacDonald, 1987), 9. Lone Fox Lake (MacDonald, 39. Goldeye Lake (Hickman and Schweger, 1993), 40. Wilcox 1987), 10. Fiddler's Pond (White and Mathewes, 1982), 11. Pass (Beaudoin and King, 1990), 41. Sunwapta Pass SP-10 Thompson (Ritchie, 1976), 12. Buffalo Narrows (Kuhry et al, 1993), (Beaudoin, 1984), 42. Sunwapta Pass SP-1 (Beaudoin, 1984), 43. 13. Mariana Lakes (Nicholson and Vitt, 1990), 14. Boone Lake3 Martens Slough (Mott and Christiansen, 1981), 44. Lake O'Hara (White and Mathewes, 1986), 15. Spring Lake (White and Mathewesi,, (Reasoner and Hickman, 1989), 45. Opabin Lake (Reasoner and 1986; Hickman and White, 1989), 16. West Naniskak Lake (Wilsoni,, Hickman, 1989), 46. Yamnuska Bog (MacDonald, 1982), 47. Wedge 1984), 17. Lake A (Wilson, 1984), 18. Cycloid Lake (Mott, 1973),, Lake (MacDonald, 1982), 48. Clearwater Lake (Mott, 1973), 49.

19. "Wood Bog" (Beaudoin, unpub.), 20. FHn Flon (Ritchie, 1976)I,I Russell (Ritchie, 1976), 50. Toboggan Lake (MacDonald, 1989), 21. Lofty Lake (Lichti-Federovich, 1970), 22. Moore Lake (Hickmani 51. Riding Mountain (E Lake) (Ritchie, 1964), 52. Chalmers Bog and Schweger, 1993), 23. Alpen Siding (Lichti-Federovich, 1972),, (Mott and Jackson, 1982), 53. Elk Valley Bog A (Hills et al, 1985), 24. Lake B (Mott, 1973), 25. Lake Isle (Hickman and Klarer, 1981)i,. 54. Chappice Lake (Vance, 1991), 55. Crestwynd (Ritchie, 1976), 26. Smallboy Lake (Vance et al., 1983), 27. Wabamun Lakes 56. Harris Lake (Sauchyn and Sauchyn, 1991), 57. Crowsnest (Hickman et al, 1984), 28. Lake A (Mott, 1973), 29. Grand Rapids Lake (Hills et ai, 1985), 58. Sewell Lake (Ritchie, 1976), 59. Hayes (Ritchie and Hadden, 1975), 30. Fairfax Lake (Hickman andiJ Lake (McAndrews, 1982), 60. Belmont Lake (Ritchie and Lichti- Schweger, 1991), 31. Muskiki Lake (Kubiw et al, 1989), 32. ExcelsioTr Federovich, 1968), 61. Glenboro (Tiger Hills) (Ritchie and Lichti- Basin (Luckman and Kearney, 1986), 33. Watchtower Basinn Federovich, 1968), 62. Lost Lake (Barnosky, 1989).

Géographie physique et Quaternaire, 49(1), 1995 86 R.E. VANCE, A.B. BEAUDOIN and B.H. LUCKMAN

Mathewes, 1986), because pine (specifically jack or (Hickman and White, 1989). Here too, the 6 ka interval lodgepole pine) regenerates vigorously in burned areas. appears transitional between low, warm-water conditions of Frequent middle Holocene forest fires were likely the prod­ the early Holocene and higher, more stable water levels uct of summers warmer and drier than today. Diatom and achieved by 4800 yr BP. Pine pollen increases at foothills pigment analyses complement the pollen stratigraphy at a sites further north (Snowshoe and Yesterday Lakes) at about third study site in the Peace River district, Spring Lake 6 ka (MacDonald, 1984, 1987) reflect the northward

200 km

FIGURE 3. Other relevant localities: Records that may span the Autres sites pertinents. Les données concernent la période de 6 6 ka interval but which generally lack firm chronologic control (A): ka, mais la chronologie n'est pas rigoureuse (A) : 63. Mari Pond 63. Marl Pond (Wilson, 1984), 64, Mary Gregg Lake (Bombin, (Wilson, 1984), 64, Mary Gregg Lake (Bombin, 1982), 65. 1982), 65. Cottonwood Slough (Kearney, 1981), 66. Callum Bog Cottonwood Slough (Kearney, 1981), 66. Callum Bog (Alley, 1972), (Alley, 1972), 67. Elkwater Lake (North, 1992). Records that gen­ 67. Elkwater Lake (North, 1992). Les données s'appliquent après erally post-date the 6 ka interval (X): 68. Reindeer Lake (Ritchie, la période 6 ka (X) : 68. Reindeer Lake (Ritchie, 1976), 69. La 1976), 69. La Ronge (Kuhry et al., 1992, 1993), 70. Baptiste Lake Ronge (Kuhry et al., 1992, 1993), 70. Baptiste Lake (Hickman et (Hickman etal., 1990), 71. Beauval (Kuhry et al., 1993), 72. Mar­ al., 1990), 71. Beauval (Kuhry et al., 1993), 72. Marguerite Lake guerite Lake (Kubiw et al., 1989), 73. Lac Ste Anne (Forbes and (Kubiw et al., 1989), 73. Lac Ste Anne (Forbes et Hickman 1981), Hickman 1981), 74. Elk Island Pond (Vance etal., 1983), Elk Island 74. Elk Island Pond (Vance et al., 1983), Elk Island (Kuhry et al., (Kuhry etal., 1993), 75. Cooking Lake (Hickman, 1987), 76. Hast­ 1993), 75. Cooking Lake (Hickman, 1987), 76. Hastings Lake (Vance ings Lake (Vance et al., 1983), 77. Gypsumville (Kuhry et al., et al., 1983), 77. Gypsumville (Kuhry et al., 1993), 78. Goat Lake 1993), 78. Goat Lake (Bujak, 1974), 79. Linnet Lake (Hills et al., (Bujak, 1974), 79. Linnet Lake (Hills et al., 1985). Les données 1985). Records that pre-date the 6 ka interval (•): 80. Herbert précèdent la période 6 ka (%).80. Herbert (Kupsch, 1960), 81. (Kupsch, 1960), 81. Hafichuk (Ritchie and De Vries, 1964), 82. Hafichuk (Ritchie et De Vries, 1964), 82. Kayville (Scrimbit) (Ritchie, Kayville (Scrimbit) (Ritchie, 1976), 83. Guardipee Lake (Bamosky, 1976), 83. Guardipee Lake (Barnosky, 1989), 84. Woodworth Pond 1989), 84. Woodworth Pond (McAndrews etal., 1967), 85. Siebold (McAndrews et al., 1967), 85. Siebold (Cvancara et al., 1971). Les (Cvancara et al., 1971). Lake records focused primarily on données lacustres concernent principalement la géochimie (U): geochemistry (•): 86. Metiskow Lake (Last, pers. comm., 1993), 86. Metiskow Lake (W.M. Last, comm . pers. , 1993), 87. Waldsea 87, Waldsea Lake (Last and Schweyen, 1985), 88. Freefight Lake Lake (Last et Schweyen, 1985), 88. Freefight Lake (Last, 1993), (Last, 1993), 89. Lake Manitoba (Last and Teller, 1983), 90. Ceylon 89. Lake Manitoba (Last et Teller, 1983), 90. Ceylon Lake (Last, Lake (Last, 1990). Other localities mentioned in the text (+): 91. 1990). Autres sites signalés dans le texte (+) : 91. Maligne Pass Maligne Pass (Luckman and Kearney, 1986), 92. Copper Lake (Luckman et Kearney, 1986), 92. Copper Lake (White et Osborn, (White and Osborn, 1992). 1992).

Géographie physique et Quaternaire. 49(1 ), 1995 THE PRAIRIE PROVINCES 87

Lofty Lake, Alberta Pollen Percentage Record

8 •2 6200 ±1401

Muama-

7480 ±1601

9180*160' i1400±190-

<1%

FIGURE 4. Lofty Lake (southern boreal forest, Fig. 2, site 21) Diagramme abrégé de pourcentages polliniques de Lofty Lake summary percentage pollen diagram (Lichti-Federovich, 1970). 6 (forêt boréale méridionale, fig. 2, site 21) (Lichti-Federovich, 1970). ka interval (5000-7000 yr BP) shaded. Mazama ash deposited L'intervalle de 6 ka (5000-7000 BP) est tramé. La cendre du mont 6850 yr BP (Bacon, 1983). Mazama a été déposée à 6850 BP (Bacon, 1983).

Holocene expansion of lodgepole pine {Pinus contorta) along suggests lower water levels than today at Toboggan Lake the eastern slopes of the Rocky Mountains (MacDonald (Fig. 5a). Kubiw et al. (1989) noted changes in peat type and Cwynar, 1985). and macrofossil assemblages at Muskiki Lake that indicate lowering of the water table between about 7500-4500 yr BP, C) ROCKY MOUNTAIN/FOOTHILLS REGION and suggested that incursions of moist Pacific at this time Pollen records from sites in the foothills and montane may have been crucial for maintaining wetlands in the foot­ forest zones of the Rockies are dominated by Pinus and hills, albeit with a lower water table than today, when sites Picea through the majority of the postglacial period. Many further east were dry. (Bear, 1989; Bombin, 1982; Hickman and Schweger, 1991, 1993; Kearney, 1981; Kearney and Luckman, 1987; In the mountains, however, the clearest indication of MacDonald, 1982, 1989; Mott and Jackson, 1982) record warmer than present 6 ka climate is the upslope migration only major vegetational changes associated with migration of the temperature-sensitive upper subalpine treeline. Pol­ of various taxa following déglaciation and the majority of len records from subalpine and alpine areas in the Alberta study sites are situated too far from major ecozone bounda­ Rockies (Beaudoin, 1986; Beaudoin and King, 1990; Kearney ries to register events driven by climatic change. Like the and Luckman, 1983; Luckman and Kearney, 1986; Reasoner boreal forest, however, the coniferous mountain vegetation and Hickman, 1989) all suggest higher than present is prone to fire, and increased fire frequency at 6 ka, com­ timberlines during the middle Holocene. For example, the pared to today, is implicated in several of these records. pollen and plant macrofossil record from Opabin Lake (Fig. Warmer and drier climatic conditions are inferred at lower 5b), a glacier-fed basin now situated above timberline in the elevations from middle Holocene maxima in Pinus pollen southern Canadian Rockies, shows an extended period of frequencies (Hills et ai, 1985; MacDonald, 1982) and peak increased Picea and Abies pollen and needle input be­ charcoal influx (Kearney and Luckman, 1987), both tween 7000 and 3000 yr BP. Clearly, treeline was posi­ byproducts of frequent fires. tioned upslope of the lake at this time in response to el­ evated summer temperatures, compared to today. Peak There is also evidence that montane and foothills diatom productivity at Opabin Lake (7000 and 5500 yr BP) wetlands may have been under considerable moisture stress supports this interpretation (Reasoner and Hickman, 1989), during the 6 ka interval, as indicated, for example, by peak because diatom proliferation in alpine lakes reflects favour­ carbonate concentration at Yamnuska Bog (MacDonald, able, warm water growing conditions. 1982). Elevated 6 ka Chenopodiineae and Cyperaceae pollen representation (compared to present), although slightly Abundant subfossil wood in alpine areas also suggests reduced from peak values achieved at about 8000 yr BP, higher than present treeline during the 6 ka interval. Luckman

Géographie physique et Quaternaire, 49(1). 1995 88 R.E. VANCE, A.B. BEAUDOIN and B.H. LUCKMAN

Toboggan Lake, Alberta I Pollen Percentage Record

3 „ < -I I 11H122 a 1 I II

Abies Alnus !I $ O. 0. OQ Ui O % O iont o- U U, U. L-, U, I "

BO

100H

160

200H

i260 6960 ±60- — •>'.•'•.'•'•>•.•

4

10400 ±70- 20 40 60 80 + - <1% FIGURE 5a) Toboggan Lake (foothills subalpine forest, Fig. 2, a) Diagramme abrégé de pourcentages polliniques de Toboggan site 50) summary percentage pollen diagram (MacDonald, 1989). Lake (forêt subalpine de piémont, fig. 2, site 50) (MacDonald, 6 ka interval (5000-7000 yr) shaded. Mazama ash deposited 6850 1989). L'intervalle de 6ka (5000-7000 BP) est tramé. La cendre BP (Bacon, 1983). du mont Mazama a été déposée à 6850 BP (Bacon, 1983).

Opabin Lake, Alberta Pollen Percentage Record o. .5

F * c » o 5 Q- ® -C a M C S £ S h o ^ Zone» 6 o o o o 10 16- 20 26 30 36 Brida* Rivti 40 46 60- 66^ 60 66 3 7Oj 6K 7B1 • H • — - 80 86 90 96- 100- 106 110J 20 40 60 80 20 40 e = mecrofossils, + = <1% FIGURE 5b) Opabin Lake (alpine tundra, Fig. 2, site 45) sum­ b) Diagramme abrégé de pourcentages polliniques de Opabin mary percentage pollen diagram, including occurrence of macro- Lake (toundra alpine, fig. 2, site 45), comprenant la présence de fossils indicative of treeline proximity. 6 ka interval (5000-7000 yr macrofossiles indicatifs de la proximité de la ligne des arbres. BP) shaded. Mazama ash deposited 6850 yr BP (Bacon, 1983); L'intervalle de 6 ka (5000-7000 BP) est tramé. La cendre du mont Bridge River ash deposited 2350 yr BP (Mathewes and Westgate, Mazama a été déposée à 6850 BP (Bacon, 1983); la cendre de 1980). Bridge River a été déposée à 2350 BP (Mathewes et Westgate, 1980).

Géographie physique et Quaternaire. 49(1), 1995 THE PRAIRIE PROVINCES 89

and Kearney (1986) report Picea logs dated ca. 7910, 8060 Athabasca Glacier but ranges in age from 7550-8230 yr BP and 8770 yr BP approximately 100 m above present treeline (Table I1 Luckman, 1988; Luckman et ai, 1993). Based on in the Watchtower Basin of the Maligne Range. Picea, the microclimate and colonization patterns in the present Abies, and Pinus needles were also recovered from sedi­ forefields of these glaciers, it seems probable that the ter­ ment older than 5960 yr BP at this site. Further south, two mini of Athabasca and Dome Glaciers were probably sev­ large Abies logs were recovered from a site 50 m above eral kilometres up-valley of their present positions during present treeline in Maligne Pass and dated to 5960 and the 6 ka interval. Warmer 6 ka summer temperatures could 5260 yr BP (Table I). These sub-fossil trees were over 3.5 also have produced higher summer runoff volumes in gla­ m long and were found with rootstocks attached, indicating cier-fed rivers that cross the prairies. Such downstream that they were buried in situ. Differences in size and ringwidth effects must be considered when evaluating characteristics between these subfossil and contemporary paleoenvironmental conditions at riparian sites along the trees (see Luckman and Kearney, 1986, Table 2) indicate major prairie rivers. that the subfossil trees were probably growing some dis­ Pollen records from lower treeline sites, particularly in tance below the contemporaneous treeline. southwestern Alberta where the transition from subalpine to Warmer, and probably drier conditions than present also grassland vegetation takes place over a relatively short allowed forest to colonize upper mountain valley floors now distance, show that coniferous forest retreated to higher occupied by wetlands. For example, subfossil trees have elevations as grassland expanded upslope at 6 ka (Hills et been recovered from two sites (Tonquin Pass and Sunwapta al., 1985; MacDonald, 1989). These ecotonal adjustments Pass) located below present treeline that are currently too suggests increased aridity compared to today, an interpre­ wet for tree growth. Stumps and logs recovered below tation that is supported by compositional changes in the Mazama tephra from sections in Tonquin Pass (Kearney local vegetation. For example, at Toboggan Lake, increased and Luckman, 1983; Luckman, 1990, Fig. 4) date to be­ amounts of Haploxylon-type pine pollen between 7600-5500 tween 4400 and 6570 yr BP Because there is little possibil­ yr BP probably represent the establishment of limber pine ity that stream or avalanche activity could have transported {Pinus flexilis) near the site (MacDonald, 1989). Limber these trees to their present location (the largest log is over pine is currently restricted to dry, rocky, lower treeline sites, 7 m long), these logs indicate that the valley floors were dry such as those along the east slopes of the Rocky Moun­ enough to support tree growth at 6 ka. Logs ranging in age tains ca. 75 km south of Toboggan Lake. Middle Holocene from 5160 to 8100 yr BP (Table I) were also recovered from increases in nonarboreal pollen frequencies indicate that peats in Sunwapta Pass. Detrital peat layers with abundant grassland expanded upslope from its current position on cones and twigs have been dated to between 8100 and the plains into floors of major valleys in the extreme south­ 6900 yr BP in two of these wetlands, again suggesting west of the study area at 6 ka (Hills et al., 1985), once somewhat drier than present conditions between 5000 and again suggesting that 6 ka climate was warmer and drier 8000 yr BP Wood dates from high elevation sites (Table I) than present. show two groupings at 5000-6600 yr BP and 7500-8800 yr D) SOUTHERN BOREAL FOREST, PARKLAND BP. These correspond broadly with the two periods of high­ Sites near the southern boreal forest margin and the est treeline position shown in the reconstructed treeline aspen parkland display the most clear-cut evidence of 6 ka curve, based on Abies/Pinus pollen ratios in the Watch- vegetation change in the western Canadian interior. In the tower Basin (Luckman and Kearney, 1986). west, many records (Hickman et al., 1984; Hickman and Recent dating of the Crowfoot Advance (Luckman and Schweger, 1993; Lichti-Federovich, 1970, 1972; Vance et Osborn 1979) to the interval between 10,100 and 11,300 yr al., 1983) show that grassland expanded north of its current BP (Reasoner et al., 1993), indicates that glaciers had limit at this time. At Lofty Lake (Lichti-Federovich, 1970), a receded, by 10,000 yr BP, to positions that were not sur­ standard for discussing the vegetation history of western passed until the Little Ice Age (ca. 600 yr BP). The infer­ Canada (Fig. 4), the 6 ka interval falls within zone L4 (7500- ence that glaciers were less extensive than today during 3500 yr BP). The distinguishing feature of L4 is increased the 6 ka interval is supported by middle Holocene sedimen­ nonarboreal pollen (especially Gramineae and to a lesser tation rate minima in glacier-fed Crowfoot Lake, because extent Artemisia, Cyperaceae, and Chenopodiineae). These low rates indicate reduced glacial sediment input (Leonard, changes suggest local expansion of open grassy areas on 1986). Recently, detrital wood, derived from subglacial xeric, south-facing slopes accompanied by movement of sources and dating to between 6100 and 6400 yr BP, has the prairie grassland margin some 80 km to the north of its current limit (Lichti-Federovich, 1970). Lofty Lake was con­ been washed out of the snout of Dome Glacier (Table I1 Luckman et al., 1993). The wood recovered includes frag­ siderably reduced in size between 8700-6300 yr BP, sug­ ments from mature, erect trees, at least 30 cm in diameter gesting that water level fell by as much as 8 m (Anderson and 200-300 years old, that was being reworked from a et al., 1989; Schweger and Hickman, 1989). This event deposit at an unknown location some distance up-valley. It probably accounts, at least in part, for the heightened indicates that ca. 6000-6400 yr BP there were mature pines Cyperaceae representation early in zone L4. growing on the valley floor or lower valley side sites of the Other sites in the southern boreal forest/parkland of Dome Valley in an area which is presently ice-covered. Alberta show that a regional drop in the water-table accom­ Similar material has been recovered from the adjacent panied the middle Holocene northward movement of

Géographie physique et Quaternaire, 49(1), 1995 90 R.E. VANCE, A.B. BEAUDOIN and B.H. LUCKMAN

TABLE I Wood older than ca. 4000 yr BP, Jasper National Park, and adjacent area

Radiocarbon date Lab. number Species/material Comments

Watchtower Basin 8770±80 GSC 3195 Plcea Logs recovered from bog 100 m above present treeline (Luckman and Kearney, 1986). 8060±90 GSC 2615 Picea 7910±70 GSC 3226 Plcea Sunwapta Pass 8100±100 GSC 2589 Picea Logs recovered from Bog 2 in Sunwapta Pass (Beaudoin, 1984) 7700±110 BGS 450 Picea 5160±60 GSC 3356 Picea 6920±100 BGS 451 Peat Detrital layer with wood and cones, Bog 4 (Beaudoin, 1984) 6850±100 AECV-324C Peat Tonquin Pass 6570±70 GSC 2468 Picea Logs or stems from small wetland in Tonquin Pass (Kearney and Luckman, 1983; Lowden and Blake, 1980) 5090±70 GSC 2631 Picea 4400±200 GSC 2927 Picea Maligne Pass 5920±70 BGS 566 Abies Logs from small wetland 50 m above present treeline (Luckman and Kearney, 1986) 5260±100 GSC 3147 Abies Athabasca Glacier 8230±80 Beta 17373 Pin us Shredded wood samples washed from Athabasca Glacier (Luckman, 1988; Luckman et ai, 1993) 8000±90 Beta 20047 IPinus 7550±100 Beta 29957 IPinus Dome Glacier 6380±80 Beta 33008 Pin us Detrital wood washed from Dome Glacier (Luckman et ai, 1993) 6120±60 Beta 33007 Pinus

AECV= Vegreville, Alberta: Beta= Beta analytic, Florida; BGS= Brock University; GSC= Geological Survey of Canada grassland. Basal dates from peatlands suggest that a low than today (Hickman et al., 1984; Anderson et ai, 1989; water-table, compared to today, inhibited development of Schweger and Hickman, 1989). Lakes that maintained per­ fen peat before 6 ka throughout much of the present south­ manent water through 6 ka were probably fed by depend­ ern boreal forest (Zoltai and Vitt, 1990). Numerous basal able, deep groundwater sources, as at Lake Wabamun radiocarbon dates on lake sediment sequences reveal that (Hickman et ai, 1984). In summary, conditions remained many basins remained dry in central Alberta at 6 ka, al­ warmer and drier than present throughout central Alberta at though infilling of some of the previously dry basins had 6 ka. begun as early as 7500 yr BP (Schweger and Hickman, To the east, a similar northward movement of grassland 1989). Deeper basins that retained water through the early at 6 ka is evident from peak nonarboreal pollen input at Holocene arid interval remained significantly reduced in scattered sites in Saskatchewan and Manitoba (Mott, 1973; size and more saline at 6 ka than they are today, an inter­ Mott and Christiansen, 1981; Ritchie, 1964, 1969, 1976; pretation based on several lines of evidence: the presence Ritchie and Lichti-Federovich. 1968; Wilson, 1984). Zone Il of Ruppia pollen (Schweger and Hickman, 1989), an aquatic pollen representation at E Lake in Riding Mountain (Ritchie, taxon that thrives in saline water (Husband and Hickman, 1964, 1966, 1969, 1983) typifies the nature of vegetation 1985); diatom assemblages suggesting reduced water lev­ changes that occurred in the southern boreal forest/park­ els (Hickman and Klarer, 1981; Hickman etal., 1984, 1990; land of these provinces (Fig. 6). Artemisia, Ambrosia, and Hickman and Schweger, 1993; Schweger and Hickman, Cheno-Am pollen frequencies are declining, but remain 1989); and peak carbonate production likely stimulated by higher than present. In contrast, Picea pollen representa­ tion is much lower than present. These changes indicate decreased water levels (Hickman and Klarer, 1981). Peak that forested uplands on Riding Mountain were occupied by charcoal input also suggests that fires were more prevalent

Géographie physique et Quaternaire. 49(1), 1995 THE PRAIRIE PROVINCES 91

Riding Mountain, Manitoba PoIUn Percentage Record

776±110a

1170±110-

1780*130»

100-

160

6026±120- E 200

6130±130« •K 1 • 260H 7280±130» O

7760 ±130«

400

11140±200« 460 20 20 20 40 20 20 40 20 40 20 <1% FIGURE 6. Riding Mountain (southern boreal forest, Fig. 2, site Diagrammme abrégé de pourcentages polliniques de Riding Moun- 51) summary percentage pollen diagram (Ritchie, 1964). 6 ka tain (forêt boréale méridionale, fig. 2, site 51) (Ritchie, 1964). interval (5000-7000 yr BP) shaded. L'intervalle de 6 ka (5000-7000 BP) est tramé.

grassland at 6 ka in response to warmer and drier climatic grassland is mainly determined by available moisture. Un­ conditions. Similar events are evident in pollen records fortunately, pollen records from the grassland region are from southern Manitoba, where grassland replaced park­ rare, and several records that have been studied are either land in the middle Holocene (Ritchie and Lichti-Federovich, truncated and restricted to the early Holocene (Ritchie and 1968), and in northwestern Ontario, where parkland re­ DeVries, 1964), or lack chronological control (Ritchie, 1976). placed boreal forest (McAndrews, 1982). Higher In addition, errors in conventional radiocarbon ages are évapotranspiration rates resulting from either higher sum­ chronic due to widespread carbonate deposits (Barnosky et mer temperatures or reduced precipitation (or both) are al., 1987; Mott,1973). However, AMS radiocarbon dating at implicated in these records on the eastern prairie margin. two recently studied sites in the driest portion of the west­ ern interior provides the first reliable chronology of Holocene As in central Alberta, middle Holocene lake levels were paleoecological events in the prairie grassland. lower than today in central Saskatchewan. Waldsea Lake, currently 14 m deep at its deepest point, was a shallow Harris Lake is a small, freshwater basin situated at lower saline playa at 6 ka (Last, pers. comm., 1993, Last and treeline on the northern flank of the Cypress Hills, a forested Schweyen, 1985). Detailed lithostratigraphic studies of nu­ enclave in southwestern Saskatchewan. Pollen and merous core samples from Lake Manitoba, one of the larg­ lithostratigraphic studies indicate that, although preceded est lakes in the prairie provinces (Fig. 1), indicate that the by a period when lake level was lower than today (9100- lake was considerably shallower during the 6 ka interval 7700 yr BP), the 6 ka interval is characterized by grassland (the lake now has a maximum depth of 6 m) and, at times, dominance of the local vegetation (Sauchyn and Sauchyn, the south basin was completely dry (Last and Teller, 1983; 1991) and mineralogical changes indicative of periodic low Teller and Last, 1990). These hydrological adjustments sug­ water interludes when lake salinity rose considerably above gest more arid conditions than today also prevailed over present values (Last and Sauchyn, 1993). However, unlike central Saskatchewan and southern Manitoba at 6 ka. many lakes in the grassland, Harris Lake did not dry up during the 6 ka interval, as indicated by the continued E) GRASSLAND occurrence Myriophyllum pollen. Paleoecological studies in the southern reaches of the Chappice Lake is a small, shallow, hypersaline lake in prairie provinces should provide a particularly sensitive the mixed-grass prairie of southeastern Alberta, 75 km north­ record of effective precipitation, since the distribution of west of Harris Lake. The Chappice Lake pollen record has

Géographie physique et Quaternaire. 49(1), 1995 92 R.E. VANCE, A.B. BEAUDOIN and B.H. LUCKMAN

limited interpretive value because it is dominated by interval, but also may have periodically attained high levels Artemisia, Gramineae, and Chenopodiineae, three taxa that (Last, 1990; Teller and Last, 1990). Guliov (1963) used include a number of species with widely varying tolerances ostracode remains to outline a period of reduced water that cannot be distinguished on the basis of pollen. Never­ levels (7000-4000 yr BP) in a small prairie pond near Regina. theless, aspects of the record suggest that the 6 ka interval Mott's (1973) pollen record from Clearwater Lake in south­ marks the transition from a lengthy period of pronounced western Saskatchewan also suggests increased 6 ka arid­ water level fluctuations (indicated by heightened Cheno-Am ity, compared to today. In contrast, near-shore sedimentary and Ambrosia percentages and fluctuating Ruppia repre­ records from Freefight Lake, a deep, hypersaline lake less sentation) to an interval of more stable, low-water, than 125 km from Clearwater Lake, suggest that the lake hypersaline conditions, as indicated by peak Ruppia fre­ was maintained at levels as high as today through the quencies (Vance, 1991; Vance et al., 1992, 1993). This middle Holocene (W.M. Last, pers. comm., 1993). Metiskow transition is also expressed in sediment type. Prior to 6000 Lake, a playa situated at the grassland/parkland boundary yr BP, massive and laminated silt and clay sequences punc­ in eastern Alberta, was a hypersaline perennial (but shal­ tuate units of well sorted silt and fine sand, the latter most low) lake during the 6 ka interval (W.M. Last, pers. comm., likely deposited during intervals of intense aeolian activity 1993). These contradictory interpretations are difficult to when the basin was dry. From 6000 to 5000 yr BP, sediments reconcile in light of previously discussed studies that sug­ are composed mainly of finely laminated carbonate-rich gest conditions drier than present were widespread through­ mud. Plant macrofossil remains (Fig. 7) are rare in pre- out much of the western Canadian interior during the mid­ 6000 yr BP sediment, a reflection of poor environment for dle Holocene. It is possible that this variability is related to preservation due to exposure and oxidation during lake local groundwater characteristics, a significant but poorly desiccation, but abundant in sediment deposited between understood aspect of prairie wetland hydrology (Last and 6000 and 5000 yr BP. Here, high Ruppia seed and Chara Slezak, 1986). Alternatively, the varied interpretations may oogonia concentrations indicate the presence of hypersaline, be more apparent than real because chronologies at most carbonate-rich waters; whereas an abundance of sites are based on conventional radiocarbon ages that could Chenopodiinaeae, Cyperaceae, and Gramineae seeds sug­ be in error by a millennium or more. It is clear however, that gest that the lake was smaller than present, although more the 6 ka interval was followed by a period of widespread productive than previously, as indicated by peak loss-on- basin infilling and lake freshening throughout much of the ignition values (Vance, 1991). southern grassland, beginning about 4000 yr BP (Vance Other grassland paleolimnological records are equivocal and Last, 1994). with regard to widespread low-water conditions during the 6 Paleoecological records from terrestrial sites are even ka interval. Sedimentological studies of numerous cores more rare than lacustrine records in the grassland, but the suggest that Ceylon Lake, a saline playa in southern Sas­ few existing studies have important implications for 6 ka katchewan, may have been dry at times through the 6 ka climate reconstructions. Bryan et al. (1987) noted numer-

Chappice Lake, Alberta ft . _ $ Plant Macrofossil Record .* Y 3 J î I I § fil I 1§ • . s I 1 « O ISI s s § i î I

enopo d 111 11Èi1S U «e iq o coccoi oaoino_i 2ones 0- • . BBfJ. BO * 100- : E=B- 160 I H U : =- 2o 200 . • • E 250- =- 300- : . t P 360 . E- 2b 400 12»0*120| I * • • • F • I =_ • FIGURE 7. Chappice Lake faa= L 2« (grassland, Fig. 2, site 54) plant U ^r — macrofossil concentration (fossils/ O ? i _^ -•r- ; ^ 100 ml) record (Vance, 1991). 6 •50 " i ' gr ka interval (5000-7000 yr BP) t^EMMMMt -J-IM ; shaded. ^^^ » i H=- f 6280*70« •• • — Concentration de plantes 800 • macrofossiles (fossiles 100/ml) à E L_ ,,, ^=- Chappice Lake (prairie, fig. 2, site •oo-l 100 200 300 SO 100 160 200 "a»" 2O 20 40 54) (Vance, 1991). L'intervalle de + = < 1 macrofossil 10OmI 6 ka (5000-7000 BP) est tramé.

Géographie physique et Quaternaire, 49(1), 1995 THE PRAIRIE PROVINCES 93

ous aeolian silt and fine sand deposits in Dinosaur Provin­ Quantitative estimates of past climatic changes based cial Park, southern Alberta. Material from the base of one of on transfer functions have been developed from two pollen these units yielded a thermoluminescence date of 5400±800 records in the southern Boreal Forest. These reconstruc­ yr BP, suggesting that the 6 ka interval was marked by tions from Riding Mountain, Manitoba (Ritchie, 1983), and increased levels of aeolian activity in southern Alberta com­ Lofty Lake, Alberta (Vance, 1986b), should be viewed with pared to today. This conclusion is supported by geological caution, since Holocene plant migrations (and other factors) and archaeological studies in the Bow River valley near can produce pollen assemblages with no known modern Calgary, where extensive loess units, deposited between analogue, a situation that undermines the entire procedure. 8000 and 5000 yr BP, have been described (Wilson, 1983; Furthermore, the transfer functions applied to the Lofty pers. com. 1993). Evidence from prairie sloughs and dug­ Lake pollen record do not reproduce modern conditions outs in Alberta (Beaudoin, 1992), Saskatchewan (Mott and accurately, particularly in terms of growing season precipi­ Christiansen, 1981) and North Dakota (Cvancara et al., tation. Bearing these caveats in mind, however, a prelimi­ 1971, McAndrews et al., 1967) indicates that freshwater nary assessment of growing season temperature and pre­ conditions prevailing immediately after déglaciation were cipitation changes at these two study sites is possible. followed by middle Holocene drying and substantial aeolian Both the Lofty Lake and Riding Mountain estimates sug­ input. Aridity and reduced vegetation cover, accompanied gest that growing season (May-August) temperature rose by high winds and likely frequent fires, appear to have rapidly ca. 10,000 yr BP and remained above present val­ stimulated considerable sediment redistribution during the ues throughout most of the Holocene. The magnitude and 6 ka interval. timing of peak temperatures vary considerably, however. Pédologie studies in southwestern Alberta river basins The Lofty Lake record indicates that growing season tem­ have revealed a period of reduced alluvial deposition and perature was about 1.50C warmer than present from 9000 soil development prior to the Mazama ash fall, indicating to 6000 yr BP, whereas the Riding Mountain record sug­ reduced flooding compared to present, presumably reflect­ gests that peak conditions were attained between 6000 and ing drier conditions (Waters and Rutter, 1984). Other inves­ 3000 yr BP, when growing season temperature exceeded tigators have reported pre-Mazama paleosols in the south­ current conditions by 60C. Although differences in the tim­ ern prairies (e.g., Kingston, 1982; Pennock and Vreeken, ing of peak Holocene warmth may in some way be a reflec­ 1986; Reeves and Dormaar, 1972) that are indicative of tion of atmospheric circulation dynamics, it is unlikely that repeated episodes of landscape stability. These features the temperature differences are realistic. Because estimates were subsequently buried by clastic influx from increased from Riding Mountain may be spuriously high due to the erosion during the 6 ka interval. Reeves and Dormaar (1972) absence of modern analogues for pollen spectra deposited suggested that a paleosol deposited ca. 8000 yr BP prob­ between 6000 and 3000 yr BP (Ritchie, 1983), but are ca. ably developed during cooler, moister conditions. Collec­ 30C above present from 8500 to 6000 yr BP, a more rea­ tively, these studies indicate that the 6 ka interval was sonable estimate of 6 ka growing season temperature is preceded by a variable climatic regime, where prevailing 1.5° to 3°C above current values. At Lofty Lake, growing warm and dry conditions were punctuated by moist inter­ season precipitation is estimated to have been 50 mm ludes. below current values between 8000 and 6000 yr BP This suggests that the early to middle Holocene aridity regis­ QUANTITATIVE ESTIMATES OF PAST CLIMATE tered in many previously discussed records was the result of longstanding above normal temperatures coupled with a Numerous records cited above indicate that the upper Holocene low in growing season precipitation. Zoltai and limit of tree growth in the Rocky Mountains increased dur­ Vitt (1990), in documenting the Holocene distribution of ing the 6 ka interval. Because treeline elevation is control­ peatlands in the Canadian western interior, arrived at slightly led mainly by temperature (Kearney, 1982), high-elevation different but comparable estimates, with mean July tem­ sites offer sensitive records of Holocene temperature perature only 0.50C warmer than today but mean annual changes that may be readily translated into estimates of precipitation 65 mm lower than present at 6000 yr BP. past temperature excursions. For example, assuming con­ temporary lapse rates, the reconstructed treeline 100 m CONCLUSION higher than present at Watchtower Basin indicates that 0 mean annual temperature was at least 0.5 C warmer than The 6 ka interval is a time of transition in the Canadian present at 6 ka (Luckman and Kearney, 1986). Most of this prairie provinces. Following late glacial-early Holocene veg­ increase is probably due to warmer growing season tem­ etation migration and landscape stabilization, many prairie peratures. Abies/Pinus pollen ratio evidence of treeline records outline an early Holocene episode of increased movement in the Excelsior Basin suggest 6 ka mean July 0 aridity between 9000 and 6000 yr BP. Although peak aridity temperatures were ca. 0.7 C warmer than present (Luckman had passed by 6000 yr BP, the landscape remained influ­ and Kearney, 1986). Oxygen isotope analyses of large Abies enced by this prolonged period of warm and dry climatic subfossil logs found above present treeline at Maligne Pass conditions (Fig. 8). In the mountains, glaciers were smaller indicate that mean annual temperatures were between 1.25° than present, upper treeline was upslope of its current and 1.60C warmer than today between 5300 and 6000 yr position, montane forests were subjected to more frequent BP (Luckman and Kearney, 1986). fires than experienced at present, and grassland persisted

Géographie physique et Quaternaire. 49(1). 1995 94 R.E. VANCE, A.B. BEAUDOIN and B.H. LUCKMAN

on presently forested floors of the major river valleys. In thei similar to present. The onset of cooler and moister condi- foothills, lower treeline moved upslope, grassland openings> tions, like the onset of postglacial warming (Ritchie and increased in size, and wetlands were reduced in size, com­• Harrison, 1993), is time-transgressive, occurring earlier pared to today. (6000-5000 yr BP) in the west and later (4000-3000 yr BP) On the plains, grassland extended northward into areas. in the east. The upper limit of tree growth in the Rocky now occupied by parkland or boreal forest and eastward in! Mountains declined markedly between 6000 and 4000 yr Manitoba. In the core of the southern prairie grassland, BP (Luckman and Kearney, 1986). Lower treeline receded many shallow lakes dried up and others were reduced toJ downslope at 5500 yr BP in western Alberta (MacDonald, hypersaline playas. Only deep lakes or those with a reliable, 1989) and, in central Alberta, lake infilling and freshening groundwater input remained perennial waterbodies. ReducedI was widespread by 5000 yr BP (Schweger and Hickman, vegetation cover and high winds led to increased aeolian! 1989). Increasingly moister conditions are evident in south­ activity, compared to today. ern Saskatchewan between 5500 yr BP (Sauchyn and Sauchyn, 1991) and 4000 yr BP (Vance and Last, 1994), The southern and northern limits of boreal forest were1 but are not apparent in southern Manitoba until about 3500 positioned further north than at present and, in the north,1 BP (Ritchie and Lichti-Federovich, 1968). trees grew in areas now occupied by tundra. Fire frequency Although quantitative estimates of the climatic changes was higher in the boreal forest than at present, aiding the responsible for these vegetation and hydrologie changes westward migration of jack pine. are varied, all concur in suggesting that increased tempera- As a time of transition, the 6 ka interval falls between the! ture was a distinguishing feature of 6 ka climate, with mean driest postglacial period and the onset of climatic conditionsi annual temperature between 0.5 and 1.5°C higher than

a b N W Ennadai Lake, I Glaciers reduced in size 2355 masl Wilcox Pass, 40 Increased lake productivity Eaglenest Lake, 5 Opabin Lake, 45 2280 masl Reduced sedimentation rates inglacier-fed lakes < > Zi: Lofty Lake, 21 Lake O'Hara, 44 • " i-^±.-r•*+•-*>-Ji-±---^T 1 E Lake, 51 2015 masl Lower lake level»; increased saliruty, more frequent fires Increased lake productivity < —> gÂT^Ui^-jllll Buffalo Lake, 38 1690 masl Maligne Lake Lower lake levels; Kettle, 36 increased salinity Lower lake levels; ?< —> increased fire Increased seofian frequency - activity i^x^j^.x^^-^r^-. ±* Chappice Lake, 54 1480 masl 1, BgSZSS Toboggan Lake, 50 E I r—I r —, ! , , , .- 7 6 5 4 Present 8 7 6 5 4 Present yrBPxlO3 yrBPxlO3

'-*: Grassland Boreal Forest Alpine Tundra •& ^Jt] Grassland/Parkland Non-analogue Parkland lllll Tundra §|i Subalpine Forest Vegetation

Non-analogue Vegetation

FIGURE 8. Summary middle Holocene palecological record for Diagramme synthèse de la paléoécologie des provinces des prai­ the Canadian prairie provinces, along a), a south-north transect ries le long d'un transect nord-sud (a) et d'un transect est-ouest en and b), an east-west elevationai transect in the foothills/Rocky progression altitudinale (b) dans la région du piémont et des Mountains region. Numbers identify site locations on Fig. 2. 6 ka Rocheuses. L'intervalle de 6 ka (5000-7000 BP) est tramé. Tous time slice (7000-5000 yr BP) shaded. All comments refer to condi­ les commentaires se rapportent à des conditions comparées à tions compared to present. celles d'aujourd'hui.

Géographie physique el Quaternaire. 49(1), 1995 THE PRAIRIE PROVINCES 95

present, and summer temperature between 0.5 and 3.O0C Barnosky, CW., 1989. Postglacial Vegetation and Climate in the North­ above current values. Reduced precipitation, compared to western Great Plains of Montana. Quaternary Research, 31: 57-73. today, is also indicated by these estimates (mean annual by Barnosky, CW.. Grimm, E.C. and Wright Jr., H.E., 1987. Towards a 65 mm and growing season by 50 mm). Postglacial History of the Northern Great Plains: A Review of the Paleoecologic Problems. Annals of Carnegie Museum, 56: 259-273. Responses to 6 ka climate outlined in prairie Bear, R.L., 1989. The Holocene Palaeoecology of Lorraine Lake, Jasper paleoecological records suggest that 6 ka atmospheric cir­ National Park, Alberta. M.Sc. dissertation, Department of Botany, Uni­ culation may have been characterized by frequent incur­ versity of Alberta. sions of Pacific air masses into the western Canadian inte­ Beaudoin, A.B., 1984. Holocene Environmental Change in the Sunwapta rior, driven by vigorous westerly flow. The zonal circulation Pass Area, Jasper National Park. Ph.D. dissertation, Department of pattern that prevailed through the 1930s (Dzerdzeevskii, Geography, University of Western Ontario, London. 1969; Namias, 1983) may be the most appropriate historic 1986. Using Picea/Pinus Ratios from the Wilcox Pass Core, Jasper analogue, with at least one significant difference. The mini­ National Park, Alberta, to Investigate Holocene Timberline Fluctuations. mal precipitation deficiencies experienced in the southern Géographie physique et Quaternaire, 40: 145-152. boreal forest in the 1930s (Chakravarti, 1976; Singh and 1992. Early Holocene Palaeoenvironmental Data Preserved in "Non- Traditional" Sites. The 2nd Palliser Triangle Global Change Confer­ Powell, 1986) are probably insufficient, even if experienced ence, Regina, Saskatchewan, Program with Abstracts. for several decades and accompanied by slightly warmer temperatures, to stimulate the widespread desiccation of 1993. A Compendium and Evaluation of Postglacial Pollen Records in Alberta. Canadian Journal of Archaeology, 17: 92-112. shallow wetlands and northward movement of grassland and boreal forest ecozones outlined in 6 ka proxy records. Beaudoin, A.B. and King, R.H., 1990. Late Quaternary Vegetation History of Wilcox Pass, Jasper National Park, Alberta. Palaeogeography, Persistence of Pacific air masses over the central area of Palaeoclimatology, Palaeoecology, 80: 129-144. the prairie provinces, produced by a northward shift of the Bombin, E.R., 1982. Holocene Paleoiimnology of Mary Gregg Lake, Foot­ westerly track compared to the 1930s, is most likely re­ hills of the Canadian Rocky Mountains, Canada. M.Sc. dissertation. quired to produce the environmental changes outlined in Department of Botany, University of Alberta, Edmonton, 147 p. this synthesis. Bryan, R.B., Campbell, I.A. and Yair, A., 1987. Postglacial Geomorphic Development of the Dinosaur Provincial Park Badlands, Alberta. Cana­ Although the effects of 6 ka climate are apparent in all dian Journal of Earth Sciences, 24: 135-146. ecozones, the clearest response to climate change is found Bryson, R.A., 1966. Air Masses, Streamlines, and the Boreal Forest. Geo­ at sensitive ecotonal sites, for example in the upper subalpine graphical Bulletin, 8: 228-269. treeline or on the fringes of the grassland. Because these Bujak, CA., 1974. Recent Palynology of Goat Lake and Lost Lake, Waterton areas provide the best opportunity to obtain numerical esti­ Lakes National Park. M.A. dissertation, Department of Geography, mates of climatic parameters for direct comparison with University of Calgary, 60 p. GCM output, ecotonal sites should be the focus of future Chakravarti, A.K., 1976. Precipitation Deficiency Patterns in the Canadian research on the Holocene environmental history of the prai­ Prairies, 1921 to 1970. Prairie Forum, 1: 95-110. rie provinces. Cvancara, A.M., Clayton, L., Bickley Jr., W.B., Jacob, A.F., Shay, CT, Delorme, L.D. and Lammers, G.E., 1971. Paleoiimnology of Late Qua­ ACKNOWLEDGEMENTS ternary Deposits: Siebold Site, North Dakota. Science, 171: 172-174, Dredge, L.A. and Cowan, W.R., 1989a. Quaternary Geology of the South­ We thank W. Johnson for providing the map that formed western Canadian Shield, p. 214-234, In R.J. Fulton, éd., Quaternary the basis of Figures 1 and 3, G.M. MacDonald and M.A. Geology of Canada and Greenland. Geology of Canada No. 1, Geologi­ cal Survey of Canada. Reasoner for permission to reproduce the Toboggan and Opabin Lakes data, K. Gajewski and H. Jette for access to 1989b. Lithostratigraphic Record on the Canadian Shield, p. 235- the Lofty Lake data, and E.C. Grimm for the Toboggan Lake 249. In R.J. Fulton, éd., Quaternary Geology of Canada and Greenland. Geology of Canada No. I Geological Survey of Canada. and Riding Mountain data. This manuscript benefitted from 1 comments by W.M. Last, R.G. Baker and an anonymous Dzerdzeevskii, B.L., 1969. Climatic Epochs in the Twentieth Century and Some Comments on the Analysis of Past Climates, p. 49-60. In H.E. reviewer. Wright Jr. éd., Quaternary Geology and Climate. Volume 16 of the Proceedings of the VII Congress of the International Association for Quaternary Science, National Academy of Sciences, Washington, D. C REFERENCES Faegri, K., Kaland, P. and Krzywinski, K., 1989. Textbook of Pollen Analy­ m Alley, N.R1 1972. The Quaternary History of Part of the Rocky Mountains, sis, 4 Edition. John Wiley, Chichester, 328 p. Foothills, Plains and Western Porcupine Hills, Southwestern Alberta. Forbes, J.R. and Hickman, H., 1981. Paleoiimnology of Two Shallow Lakes Ph.D. dissertation, Department of Geography, University of Calgary. in Central Alberta, Canada. Internationale Revue der gesamten Anderson, T.W., Mathewes, R.W. and Schweger, CE., 1989. Holocene Hydrobiologie, 66: 863-877. Climatic Trends in Canada with Special Reference to the Hypsithermal Guliov, P., 1963. Paleoecology of Invertebrate Fauna from Post-glacial Interval, p. 520-528. In R.J. Fulton, éd., Quaternary Geology of Canada Sediment Near Earl Grey, Saskatchewan. M.A. thesis. Department of and Greenland. Geology of Canada No. 1. Geological Survey of Canada. Geological Sciences. University of Saskatchewan, Saskatoon, 95 p. Atmospheric Environment Service, 1982. Canadian Climate Normals 1951- Hickman, M., 1987. Paleoiimnology of a Large Shallow Lake, Cooking 1980. Environment Canada, Ottawa. Lake, Alberta, Canada, Archiv fur Hydrobiologie, 111: 121-136. Bacon, CR., 1983. Eruptive History of Mount Mazama and Crater Lake Hickman, M. and Klarer, D.M., 1981. Paleoiimnology of Lake Isle, Alberta, Caldera, Cascade Range, U.S.A. Journal of Volcanology and Geothermal Canada (Including Sediment Chemistry, Pigments and Diatom Research, 18: 57-115. Stratigraphy). Archiv fur Hydrobiologie, 91: 490-508.

Géographie physique el Quaternaire. 49(1), 1995 96 R.E. VANCE, A.B. BEAUDOIN and B.H. LUCKMAN

Hickman, M. and Schweger, CE., 1991. A Palaeoenvironmental Study of 1993. Rates of Sediment Deposition in a Hypersaline Lake in the Fairfax Lake, a Small Lake Situated in the Rocky Mountain Foothills of Northern Great Plains, Western Canada. International Journal of Salt West-central Alberta. Journal of Paleolimnology, 6: 1-15. Research, 2: 47-58. 1993. Late Glacial-Early Holocene Paleosalinity in Alberta, Canada Last, W.M. and Sauchyn, D.J., 1993. Mineralogy and Lithostratigraphy of — Climate Implications. Journal of Paleolimnology, 8: 149-161. Harris Lake, Southwestern Saskatchewan, Canada. Journal of Paleolimnology, 9: 23-39. Hickman, M., Schweger, CE. and Habgood, T, 1984. Lake Wabamun, AJta.: A Paleoenvironmental Study. Canadian Journal of Botany, 62: Last, W.M. and Schweyen, TH., 1985. Late Holocene History of Waldsea 1438-1465. Lake, Saskatchewan, Canada. Quaternary Research, 24: 219-234. Hickman, M., Schweger, CE. and Klarer, D.M., 1990. Baptiste Lake, Al­ Last. W. and Slezak, L.A., 1986. Paleohydrology, Sedimentology, and berta — A Late Holocene History of Changes in a Lake and its Catch­ Geochemistry of Two Meromictic Saline Lakes in Southern Saskatch­ ment in the Southern Boreal Forest. Journal of Paleolimnology, 4: 253- ewan. Géographie physique et Quaternaire, 40: 5-15. 267. Last, W.M. and Teller, J.T., 1983. Holocene Climate and Hydrology of the Hickman, M. and White, J.M., 1989. Late Quaternary Paleoenvironments of Lake Manitoba Basin, p. 333-353. In JT. Teller, and L. Clayton, eds., Spring Lake, Alberta, Canada. Journal of Paleolimnology, 2: 305-317. Glacial Lake Agassiz. Geological Association of Canada Special Paper No. 26. Hills, L.V., Christensen, O.A., Fergusson, A., Driver, JC. and Reeves, B.O.K., 1985. Postglacial Pollen and Paleoclimate in Southwestern Laycock, A.H., 1972. The Diversity of the Physical Landscape, p. 1-32. In Alberta and Southeastern Saskatchewan, p. 345-396. In CR. Harington, P.J. Smith, éd., Studies in Canadian Geography — The Prairie Prov­ éd., Climatic Change in Canada 5. Critical Periods in the Quaternary inces. The University of Toronto Press. Climatic History of Northern North America. National Museum of Natu­ Leonard, E.M., 1986. Use of Lacustrine Sedimentary Sequences as Indica­ ral Sciences Project on Climatic Change in Canada During the Past tors of Holocene Glacial History, Banff National Park, Alberta, Canada. 20,000 Years. Syllogeus No. 55. Quaternary Research, 26: 218-231. Husband, B.C. and Hickman, M., 1985. Growth and Biomass Allocation of Lichti-Federovich, S., 1970. The Pollen Stratigraphy of a Dated Section of Ruppia occidentalis in Three Lakes, Differing in Salinity. Canadian Late Pleistocene Lake Sediment Core from Central Alberta. Canadian Journal of Botany, 63: 2004-2014. Journal of Earth Sciences, 7: 938-945. Janz, B. and Storr, D., 1977. The Climate of the Contiguous Mountain 1972. Pollen Stratigraphy of a Sediment Core from Alpen Siding Parks. Project Report 30, prepared for Parks Canada. Atmospheric Lake, Alberta. Report of Activities, Part B. Geological Survey of Canada Environment Service, Department of Environment, 324 p. Paper, 72-1B: 113-115. Jette, H., 1995. A Canadian Contribution to the Paleoclimate Model Longley, R.W., 1972. The Climate of the Prairie Provinces. Climatological lntercomparison Project (PMIP). Géographie physique et Quaternaire, Studies No. 13. Environment Canada, Atmospheric Environment Serv­ 49 (1): 4-12. ice. Kearney, M.S., 1981. Late Quaternary Vegetational and Environmental Lowden, J.A. and Blake Jr., W., 1980. Geological Survey of Canada, History of Jasper National Park, Alberta. Ph.D. dissertation. Depart­ Radiocarbon Dates XIX. Geological Survey of Canada Paper 79-7. ment of Geography, University of Western Ontario, London, 318 p. Luckman, B.H., 1988. 8000 Year Old Wood from the Athabasca Glacier, 1982. Recent Seedling Establishment at Timberline in Jasper Na­ Alberta. Canadian Journal of Earth Sciences, 25: 148-151. tional Park, Alta. Canadian Journal of Botany, 60: 2283-2287. 1990. Mountain Areas and Global Change: A View from the Cana­ Kearney, M.S. and Luckman, B.H., 1983. Postglacial Vegetational History dian Rockies. Mountain Research and Development, 10: 183-195. of Tonquin Pass, British Columbia. Canadian Journal of Earth Sciences, Luckman, B.H., Holdsworth, G. and Osborn, G.D., 1993. Neoglacial Glacial 20: 776-786. Fluctuations on the Canadian Rockies. Quaternary Research, 39: 144- 1987. A Mid-Holocene Vegetational and Climatic Record from the 153. Subalpine Zone of the Maligne Valley, Jasper National Park, Alberta Luckman, B.H. and Kearney, M.S., 1986. Reconstruction of Holocene (Canada). Palaeogeography, Palaeoclimatology, Palaeoecology, 59: 227- Changes in Alpine Vegetation and Climate in the Maligne Range, Jas­ 242. per National Park, Alberta. Quaternary Research, 26: 244-261. Kingston, M.S., 1982. An Evaluation of the Recommended Methods for Luckman, B.H. and Osborn, G.D., 1979. Holocene Glacier Fluctuations in Paleosol Investigations; A Case Study of a Holocene Paleosol, Calgary, the Middle Canadian Rocky Mountains. Quaternary Research, 11: 52- Alberta. M.Sc. dissertation, Department of Geography, University of 77. Western Ontario, London, 166 p. MacDonald, G.M., 1982. Late Quaternary Paleoenvironments of the Morley Klassen, R.W., 1989. Quaternary Geology of the Southern Canadian Inte­ Flats and Kananaskis Valley of Southwestern Alberta. Canadian Jour­ rior Plains, p. 138-173. In RJ. Fulton, éd., Quaternary Geology of nal of Earth Sciences, 19: 23-35. Canada and Greenland. Geology of Canada No. 1. Geological Survey 1984. Postglacial Plant Migration and Vegetation Development in of Canada. the Western Canadian Boreal Forest. Ph.D. dissertation. Department of Kubiw, H., Hickman, M. and Vitt, D.H., 1989. The Development of Peatlands Botany, University of Toronto. at Muskiki and Marguerite Lakes, Alberta. Canadian Journal of Botany, 1987. Postglacial Development of the Subalpine-Boreal Transition 67: 3534-3544. Forest of Western Canada. Journal of Ecology, 75: 303-320. Kuhry, P., Halsey, L.A., Bayley, S.E. and Vitt, D.H., 1992. Peatland Devel­ 1989. Postglacial Palaeoecology of the Subalpine Forest-Grassland opment in Relation to Holocene Climatic Change in Manitoba and Ecotone of Southwestern Alberta: New Insights on Vegetation and Saskatchewan (Canada). Canadian Journal of Botany, 29: 1070-1090. Climate Change in the Canadian Rocky Mountains and Adjacent Foot­ Kuhry, P., Nicholson, Gignac, L.D., Vitt, D.H. and Bayley, S.E., 1993. hills. Palaeogeography, Palaeoclimatology, Palaeoecology, 73: 155-173. Development of Sphagnum-dominated Peatlands in Boreal Continental MacDonald, G.M., Beukens, R.P. and Kieser, W.E., 1991. Radiocarbon Canada. Canadian Journal of Botany, 71: 10-22. Dating of Limnic Sediments: A Comparative Analysis and Discussion. Kupsch, W.O., 1960. Radiocarbon-dated Organic Sediment Near Herbert, Ecology, 72: 1150-1155. Saskatchewan. American Journal of Science, 258: 282-292. MacDonald, G.M., Beukens, R.P. Kieser, W.E. and Vitt, D.H., 1987. Com­ Last, W.M., 1990. Paleochemistry and Paleohydrology of Ceylon Lake, a parative Radiocarbon Dating of Terrestrial Plant Macrofossils and Aquatic Salt-Dominated Playa Basin in the Northern Great Plains, Canada. Moss from the "Ice-Free Corridor" of Western Canada. Geology, 15: 837- Journal of Paleolimnology, 4: 219-238. 840.

Géographie physique et Quaternaire. 49(1), 1995 THE PRAIRIE PROVINCES 97

MacDonald, G.M. and Cwynar, L.C., 1985. A Fossil Pollen Based Recon­ 1966. Aspects of the Late-Pleistocene History of the Canadian struction of the Late Quaternary History of Lodgepole Pine (Pinus Flora, p. 66-80. In R.L. Taylor, and R.A. Ludwig, eds., The Evolution of contorta ssp. Iatifolia) in the Western Interior of Canada. Canadian Canada's Flora. University of Toronto Press. Journal of Forest Research, 15: 1039-1044. 1969. Absolute Pollen Frequency and Carbon-14 Age of a Section MacDonald, G.M. and Ritchie, J.C, 1986. Modern pollen spectra from the of Holocene Lake Sediment from the Riding Mountain Area of Mani­ western interior of Canada and the interpretation of late Quaternary toba. Canadian Journal of Botany, 47: 1345-1349. vegetation development. New Phytologist, 103: 245-268. 1976. The Late-Quaternary Vegetational History of the Western Mathewes, R.W. and Westgate, J.A. 1980. Bridge River Tephra: Revised Interior of Canada. Canadian Journal of Botany, 54: 1793-1818. Distribution and Significance for Detecting Old Carbon Errors in Radio­ 1983. The Paleoecology of the Central and Northern Parts of the carbon Dates of Limnic Sediment in Southern British Columbia. Cana­ Glacial Lake Agassiz Basin, p. 157-170. InJJ. Teller, and L. Clayton, dian Journal of Earth Sciences, 17: 1454-1461. éd., Glacial Lake Agassiz. Geological Association of Canada Special Matthews Jr., J.V., 1980. Paleoecology of John Klondike Bog, Fisherman Paper No. 26. Lake Region, Southwest District of Mackenzie. Geological Survey of 1985. Quaternary Pollen Records from the Western Interior and the Canada Paper, 80-22. 12 p. Arctic of Canada, p. 327-352. In V.M. Bryant Jr. and R.G. Holloway, McAndrews, J.H., 1982. Holocene Environment of a Fossil Bison from eds., Pollen Records of Late-Quaternary North American Sediments. Kenora, Ontario. Ontario Archaeology, 37:41-51. AASP Foundation, Texas. McAndrews, J.H., Stewart Jr., R.E. and Bright, R.C., 1967. Paleoecology of 1987. Postglacial Vegetation of Canada. Cambridge University Press, a Prairie Pothole: A Preliminary Report, p. 101-113. In L Clayton and 178 p. T.F. Freers, eds., Glacial Geology of the Missouri Couteau and Adjacent 1989. History of the Boreal Forest in Canada, p. 508-512. In R.J. Areas. Miscellaneous Series 30. North Dakota Geological Survey, Grand Fulton, éd., Quaternary Geology of Canada and Greenland. Geology of Forks. Canada No. 1. Geological Survey of Canada. Moore, P.D., Webb, J.A. and Collinson, M.E., 1991. Pollen Analysis, 2™ Ritchie, J.C. and de Vries, B., 1964. Contributions to the Holocene Edition. Blackwell , Oxford, 216 p. Paleoecology of West Central Canada. II. A Late-glacial Deposit from Mo», R.J., 1973. Palynological Studies in Central Saskatchewan. Pollen the Missouri Couteau. Canadian Journal of Botany, 42: 677-692. Stratigraphy from Lake Sediment Sequences. Geological Survey of Ritchie, J.C. and Hadden, K.A., 1975. Pollen Stratigraphy of Holocene Canada Paper, 72-49. Sediments from the Grand Rapids Area, Manitoba. Review of 1978. Populus in late-Pleistocene pollen spectra. Canadian Journal Palaeobotany and Palynology, 19: 193-202. of Botany, 56: 1021-1031. Ritchie, J.C. and Harrison, S.P., 1993. Vegetation, Lake Levels, and Cli­ Mott, R.J. and Christiansen, E.A., 1981. Palynological Study of Slough mate in Western Canada during the Holocene, p. 401-414. In H.E. Sediments from Central Saskatchewan. In Current Research B. Geo­ Wright, Jr., J.E. Kutzbach, T Webb III, W.F. Ruddiman, FA. Street- logical Survey of Canada Paper, 81-1B: 133-136. Perrott and PJ. Bartlein, eds., Global Climates Since the Last Glacial Maximum. University of Minnesota Press, Minneapolis Mott, R.J. and Jackson Jr., LE., 1982. An 18,000 Year Palynological Record from the Southern Alberta Segment of the Classical Wisconsinan Ritchie, J.C. and Lichti-Federovich, S., 1968. Holocene Pollen Assem­ "Ice Free Corridor". Canadian Journal of Earth Sciences, 19: 504-513. blages from the Tiger Hills, Manitoba. Canadian Journal of Earth Sci­ ences, 5: 873-880. Namias, J., 1983. Some Causes of United States Drought. Journal of Climate and Applied Meteorology, 22: 30-39. Rowe, J.S., 1972. Forest Regions of Canada. Canadian Forestry Service, Publication No. 1300. Department of Environment, 172 p. North, M.A., 1992. Preliminary Palynological Record from Elkwater Lake, Cypress Hills, Alberta, Covering the Last 5,000 years. The 2°° Palliser Sarna-Wojcicki, A.M., Champion, D.E. and Davis, J.O., 1983. Holocene Triangle Global Change Conference, Program with Abstracts. Volcanism in the Conterminous United States and the Role of Silicic Volcanic Ash Layers in Correlation of Latest-Pleistocene and Holocene Nichols, H., 1967, The Post-glacial History of Vegetation and Climate at Deposits, p. 52-77. In H.E. Wright Jr., éd., Late Quaternary Environ­ Ennadai Lake, Keewatin, and Lynn Lake, Manitoba, Canada. Eiszeitalter ments of the United States Volume 2. The Holocene. University of und Gegenwart, 18: 176-197. Minnesota Press, Minneapolis. 1969. The Late Quaternary History of Vegetation and Climate at Sauchyn, M.A. and Sauchyn, D.J., 1991. A Continuous Record of Holocene Porcupine Mountain and Clearwater Bog. Arctic and Alpine Research, Pollen from Harris Lake, Southwestern Saskatchewan, Canada. 1: 155-167. Palaeogeography, Palaeoclimatology, Palaeoecology, 88: 13-23. Nicholson, B. and Vitt, D.H., 1990. The Paleoecology of a Peatland Com­ Schweger, CE. and Hickman, M., 1989. Holocene Paleohydrology, Central plex in Continental Western Canada. Canadian Journal of Botany, 68: Alberta: Testing the General-Circulation-Model Climate Simulations. 121-138. Canadian Journal of Earth Sciences, 26: 1826-1833. Pennock, DJ. and Vreeken, W.J., 1986. Influence of Site Topography on Singh, T and Powell, J.M., 1986. Climatic Variation and Trends in the Paleosol Formation in the Highwood River Basin, Southern Alberta. Boreal Forest Region of Western Canada. Climatic Change, 8: 267- Canadian Journal of Soil Science, 66: 673-688. 278. Reasoner, M.A. and Hickman, M., 1989. Late Quaternary Environmental Stuiver, M. and Becker, B., 1993. High-Precision Decadal Calibration of the Change in the Lake O'Hara Region, Yoho National Park, British Colum­ bia. Palaeogeography, Palaeoclimatology, Palaeoecology, 72: 291-316. Radiocarbon Time Scale, AD 1950-6000 BP. Radiocarbon, 35: 35-65. Teller, JT. and Last, W.M., 1990. Paleohydrological Indicators in Playas in Reasoner, M.A., Osborn, G. and Rutter, N.W., 1993. The Age of the Crowfoot Salt Lakes, With Examples from Canada, Australia, and Africa. Moraine System in the Canadian Rocky Mountains: A Possible Younger Palaeogeography, Palaeoclimatology, Palaeoecology, 76: 215-240. Dryas Time Equivalent Glacial Event. CANQUA '93 Program with Ab­ stracts and Field Guide, p. A37. Vance, R.E., 1986a. Pollen Stratigraphy of Eaglenest Lake, Northeastern Alberta. Canadian Journal of Earth Sciences, 23: 11-20. Reeves, B.O.K. and Dormaar, J., 1972. A Partial Holocene Pedological and Archaeological Record from the Southern Alberta Rocky Mountains. 1986b. Aspects of the Postglacial Climate of Alberta: Calibration of Arctic and Alpine Research, 4: 325-336. the Pollen Record. Géographie physique et Quaternaire, 40: 153-160. Ritchie, J.C, 1964. Contributions to the Holocene Paleoecology of West- 1991. A Paleobotanical Study of Holocene Drought Frequency in Central Canada. I. The Riding Mountain Area. Canadian Journal of Southern Alberta. Ph.D. dissertation. Department of Biological Sci­ Botany, 42: 181-197. ences, Simon Fraser University, Burnaby, 180 p.

Géographie physique el Quaternaire, 49(1), 1995 98 R.E. VANCE, A.B. BEAUDOIN and B.H. LUCKMAN

Vance, R.E., Clague, J.J. and Mathewes, R.W., 1993. Holocene 1986. Postglacial Vegetation and Climatic Change in the Upper Paleohydrology of a Hypersaline Lake in Southeastern Alberta. Journal Peace River District, Alberta. Canadian Journal of Botany, 64: 2305- of Paleolimnology, 8: 103-120. 2318. White, J.M., Mathewes, R.W. and Mathews, W.H., 1985. Late Pleistocene Vance, R.E., Emerson, D. and Habgood, T., 1983. A Mid-Holocene Record Chronology and Environment of the "Ice-Free Corridor" of Northwest- of Vegetative Change in Central Alberta. Canadian Journal of Earth em Alberta. Quaternary Research, 24: 173-186. Sciences, 20: 364-376. White, J.M. and Osborn, G., 1992. Evidence for a Mazama-like Tephra Vance, R.E. and Last, W., 1994. Paleolimnology and Global Change on the Deposited ca. 10 000 BP at Copper Lake, Banff National Park, Alberta. Canadian Prairies. In Current Research, 1994-B: 49-58. Geological Canadian Journal of Earth Sciences, 29: 52-62. Survey of Canada, Ottawa. Wilson, M.A., 1984. Postglacial Climatic and Vegetation History of the La Vance, R.E. Mathewes, R.W. and Clague, J.J., 1992. 7000 Year Record of Ronge Area, Northern Saskatchewan. SRC Publication No. R-743-2-E- 84. Saskatchewan Research Council, Saskatoon, 90 p. Lake-Level Change on the Northern Great Plains: A High-Resolution Proxy of Past Climate. Geology, 20: 879-882. Wilson, M.C, 1983. Once Upon A River: Archaeology and Geology of the Bow River Valley at Calgary, Alberta, Canada. National Museum of Man Waters, PL. and Rutter, N.W., 1984. Utilizing Paleosols and Volcanic Ash in Mercury Series. Archaeological Survey of Canada, Paper No. 114, Correlating Holocene Deposits in Southern Alberta, p. 203-223. In W.C. Ottawa. 464 p. Mahaney, éd., Correlation of Quaternary Chronologies. GeoBooks, Zoltai, S.C, 1989. Late Quaternary Volcanic Ash in the Peatlands of Cen­ Norwich. tral Alberta. Canadian Journal of Earth Sciences, 26: 207-214. White, J.M. and Mathewes, R.W., 1982. Holocene Vegetation and Climatic Zoltai, S.C. and Vitt, D.H., 1990. Holocene Climatic Change and the Distri­ Change in the Peace River District, Canada. Canadian Journal of Earth bution of Peatlands in Western Interior Canada. Quaternary Research, Sciences, 19: 555-570. 33: 231-240.

Géographie physique el Quaternaire, 49(1), 1995